| Glume fertility in primitive bread wheats of the Indian
subcontinent G. M. WRIGHT Crop Research Division, Department of Scientific and Industrial Research, Lincoln, New Zealand A wheat spikelet is said to have a "fertile" glume if a flower or some flower part, however rudimentary, is present in the axil of the upper glume (WRIGHT 1969). The level of glume fertility is strongly inherited, and for most purposes is adequately measured as the proportion of terminal spikelets with fertile glumes. Variation in glume fertility among primitive bread wheats may be related to their evolutionary history. At least, glume fertility has been showed to be lowest in wheats from high altitudes, and in wheats from countries furthest from South-west Asia, and it has been suggested that the wheats with fertile glumes are the more recently synthesised (WRIGHT 1972). No such pattern was established in Triticum turgidum, in which high levels of glume fertility are rare. None of the Pakistani wheats in the Crop Research Division collection, and few from India, were included in the previous study. The majority were collected in the mid 1920's for the British Board of Trade, and information on their places of origin, and in some cases names, has been obtained recently from the Plant Breeding Institute, Cambridge, England. The wheats from India, Pakistan, Nepal, Sikkim, and Bhutan were scored for glume fertility in 1971-72. Those with fertile glumes from India which had also been recorded in 1969-70 averaged 48.4 and 47.9 percent of fertile terminal-spikelet upper glumes respectively, so that no seasonal adjustment was required, and counts of all wheats scored in both seasons were combined. If no fertile glumes are found in 10 ears no further ears are taken ; otherwise 10 more ears are examined, and the line scores 1 with 1 to 4 fertile terminal spikelets, 2 with 5 to 9 and 3 with 10 to 20. In 7 percent of the wheats fewer than 20 ears were available for recording Among the major wheat-growing regions of India Gujarat, Maharashtra, and Rajasthan are not represented. The numbers of wheats from most regions are low, with at best perhaps one line per 7,000 hectares of wheat from Kashmir and one per 50,000 hectares from Uttar Pradesh (cf. PAL 1966, p.8), and there is probably some duplication and misidentification of lines. Among the named varieties, however, there were 9 accessions of the widely-grown Pusa wheat N.p. 4, of which only one had been identified before scoring. These had similar herbarium specimens and gave similar scores for glume fertility (0 or 1), but one line as now grown has tip awns instead of being awnless. There were several duplicates of other pure line selections, and in some of the few cases where these were different a description in PAL (1966) was available to identify the correct line. A different situation was shown by 5 lines of N.P. 12, which were similar in appearance but had more than chance variation in glume fertility. This variety appears twice in the Bihar wheats of Table 1, scoring 0 and 2. The summary of scores in Table 1 shows some of the pattern of variation over the subcontinent. There is a rather irregular decline in glume fertility from central to North-east India, with very low, though imprecise, averages in Northern Bihar and the Himalayan foothills. The wheats from the undivided Punjab and Kashmir are in line with these, inasmuch as they show a decline with increasing altitude. The regions or dis tricts in Baluchistan relate quite well to accessibility from the coast, and the figures for Sibi and Loralai, districts near the Bolan Pass, are close to the 1.54 found in the adjacent Kandahar region of Afghanistan (WRIGHT 1972). The Chagai region appears to be as isolated as Kashmir. |
| --> Next |